Abstract
Objectives:
Anterior cruciate ligament reconstruction (ACLR) has traditionally been reserved for younger, more active patients that sustain ACL injuries. Since life expectancy and healthcare delivery have been improving, more patients have been wanting to maintain active lifestyles and continue participating in sports at older age. There has thus been an interest in studying ACLR outcomes for this older patient cohort, particularly patients aged 40 and older. Although these patients may receive nonoperative treatment more frequently, prior literature has studied ACLR outcomes in this population, yet these studies are limited in size and granularity. Despite the fact that patient-reported outcome measures (PROMs) are being increasingly used in orthopaedics to define successful treatment and even influence prior authorization and reimbursement policies, PROM data is scarce for this population. Conventional thought states that surgical technique varies for this patient population compared to younger patients particularly regarding graft type, with surgeons opting for allograft use more often in older patients. Few studies have assessed the use of allograft versus autograft in these patients, with almost no studies reporting on PROMs. To the knowledge of the authors, no study has evaluated ACLR outcomes using the Knee Injury & Osteoarthritis Outcome Score (KOOS), a validated PROM commonly used for ACLR. The purpose of this study was to compare baseline factors, surgical characteristics, and outcomes between patients aged 40 and over that received allograft and those that received autograft for primary ACLR. The secondary aim was to report on available KOOS data for these patients. The hypothesis was that patients receiving allograft and autograft would demonstrate similar outcomes.
Methods:
This was a retrospective cohort study conducted at a single ambulatory surgical center in a metropolitan area. The institutional electronic medical record (EMR) was queried for ACLRs from 2009 and 2016 using CPT code terminology for ACLR and related procedures. Patients aged younger than 40, revision procedures, and concomitant ligaments reconstructions/repairs were excluded. Demographic, injury, and surgical characteristics were extracted via EMR chart review. Costs regarding implants, day of surgery, and the two-year episode of care were determined using time-driven activity-based costing.
Clinical outcomes included graft failure/re-rupture and re-operations within two years after index ACLR. The institutional PROM database was queried for available KOOS data at minimum two-year follow-up and cross-referenced with the initial patient list. Single Assessment Numeric Evaluation (SANE) was also included. Chi-square, Fisher exact, and student t-tests were conducted to identify group differences.
Results:
A total of 141 patients were included. The cohort was mostly female (57.4%) with an average age of 47 (range: 40 to 66). Average BMI was 26.5, and most patients (72.6%) were ASA score 1. Most patients had acute injuries under six months (90.1%) and had some degree of pre-existing osteoarthritis (61.0%), although this was mostly mild and almost exclusively patellofemoral. Most patients were treated with allograft (67.4%), followed by hamstring autograft (17.7%) and bone-tendon-bone autograft (14.9%). Approximately 58.2% injured at least one meniscus, with 89.0% of these receiving operations consisting of almost exclusively meniscectomies (89.0% of operations). ACLR failure and re- operation rates at two years were relatively low at 4.3% and 7.8%, respectively. Only 123 patients (87.2%) and 77 patients (54.6%) reported SANE and KOOS data at two-year follow-up, respectively. Average baseline and two-year KOOS were 59.2 and 80.0, respectively, corresponding to an average two-year improvement of +20.8. Baseline and two-year SANE were 61.2 and 85.3, respectively, corresponding to an average improvement of +23.3. Between patients receiving allograft and patients receiving autograft, there were no differences in re-rupture or re-operation rates, two-year KOOS, or two-year SANE. Statistically significant differences (allograft vs autograft) were identified for age (48 vs 45 years old, p=0.0037), operative time (91 vs 105 minutes, p=0.0053), selection of screw vs suspensory fixation in the femur (86.3% vs 47.8%, p<0.0001), femoral and tibial implant costs ($158.55 vs $247.12, p<0.0001), and overall surgical and two-year costs ($5715.90 vs $3235.42 and $6480.63 vs $4017.61, respectively, p<0.0001). More patients in the allograft group had pre-existing arthritis, but this did not reach significance (66.3% vs 50.0%, p=0.0626).
Conclusions:
Primary ACLR for patients aged 40 and older was shown to provide favorable clinical outcomes and PROs. This is one of the largest existing samples for this patient cohort and is the first study to the knowledge of the authors to provide KOOS data for this older patient population receiving ACLR, which demonstrated on average notable improvements at two years. Although most patients received allograft, outcomes did not differ between these patients and those receiving autografts. Although allograft ACLRs exhibited lower operative time and choice of less costly implants, these did not outweigh the surgical and two-year costs associated with allograft use versus native tissue harvest and did not lead to improved PROs. The interplay of these variables, in addition to presence and degree of pre-existing arthritis, on graft choice and outcomes may need to be further elucidated with studies of higher sample size and PROM response rate. Nonetheless, these results agree with prior literature and suggest that primary ACLR in the context of patients aged 40 and older provides good outcomes regardless of graft type used.
